University of minnesota Duluth

 

 Laboratory for Spatial Analysis

in

The GeoSciences

 


 

Cass county - Ten Mile Lake association Water resource management tools

 


September 30, 2003

 

 

 

CASS COUNTY - TEN MILE LAKE ASSOCIATION WATER RESOURCE MANAGEMENT TOOLS

 

1.       Groundwater Susceptibility Analysis (GWSA)

2.       Surface Water Susceptibility Analysis (SWSA)

3.       Rational Method Runoff Calculator (RMRC)

4.       Aquifer Probability Coverage (APC)

5.       Groundwater Flow Model (GWFM)

6.       Water Budget Analysis (WBA)

 

 

 

 

September 2003

 

 

 

Prepared for:

Cass County

And

Ten-Mile Lake Association

 

Prepared by:

University of Minnesota Duluth

Laboratory for Spatial Analysis in the Geosciences (LSAG)

 

 

 

Executive Summary

  This report summarized the results of an analysis of surface water and groundwater in the Ten Mile Lake watershed and surrounding area.  Included with this report are a group of tools to aid the Ten Mile Lake Association and their cooperators in managing water resources for land-use planning.  There are six major components: groundwater susceptibility analysis (GWSA), surface water susceptibility analysis (SWSA), rational method runoff calculator (RMRC), a 3D aquifer probability coverage (APC), Groundwater flow model (GWFM), and a water budget for Ten Mile and Birch Lakes (WBA).

The GWSA is based upon a model of aquifer sensitivity to ground water contamination from pollutants introduced at or near the surface.  The model was built using DRASTIC: A standardized system for evaluating ground water pollution potential using hydrogeologic settings.  Inputs to the DRASTIC model consist of seven parameters that define the intrinsic characteristics of the hydrogeologic system including depth to water, recharge, aquifer media, soils, topography, impact of vadose zone, and hydraulic conductivity of the aquifer.  The model output consists of a grid coverage containing relative aquifer sensitivity rankings for the Ten Mile Lake Watershed and surrounding area.  The rankings are classified into four categories of sensitivity: 1) low, 2) moderate, 3) high, and 4) very high based on the results of the analysis. 

The Surface Water Susceptibility Analysis (SWSA) determines relative runoff potential based on inputs of distance to water bodies (major rivers, lakes and streams), slope, land cover and soil parameters.

The Rational Method Runoff Calculator (RMRC) is an extension developed by UMD using Arcview Version 3.2.  The purpose of RMRC is to calculate the peak discharge (Q) utilizing the Rational Method from a user-defined watershed.  This is done in conjunction with another extension named “Basin1” which uses digital elevation data to derive watersheds based on a user-defined point.

The Aquifer Probability Coverage is derived from the water well drillers logs housed in the Minnesota County Well Index.  Stratigraphic information is extracted using a binary indicator.  If a stratigraphic unit is interpreted to be an aquifer it is given a value of 1, whereas units interpreted to be aquitards are given values of 0.  The water well data are then converted the locations of the wells, the material depths, and the binary indicator are converted to an x, y, z data file.  Geostatistical analysis is used to interpreted among wells to construct a 3D data set that reflects the probability that any particular location and elevation is an aquifer.  Numbers near 1 are likely aquifers and numbers near zero are likely aquitards.  The data used as input to groundwater flow models and are incorporated into an Arc/Info 3D ASCII grid file for import to 3D analyst in Arc/View or Arc/Info.

The Groundwater Flow Model is a numerical simulation of groundwater flow in the Ten Mile Lake Watershed and surrounding area.  The flow model was developed using MODFLOW with GMS (Groundwater Modeling System) as the data pre and post-processor.  The model is a steady-state representation of the groundwater system, which can be used to determine flow direction, flow rates, and to delineate wellhead protection areas. 

The Water Budget Analysis is an assessment of the hydrologic inputs and outputs to the Ten Mile and Birch Lake Watersheds.  The water budgets were compiled on an annual basis for the years 2000, 2001, and 2002.  The goal was to examine the magnitude of water exchange between Tenmile and Birch Lakes for use in managing lake levels and understanding surface and subsurface exchange of water in the complex topographic setting of the Ten Mile Lake area.

 


Table of Contents

1.0       Introduction. 8

1.1      Project Location. 8

1.2      Project Personnel 9

1.3      Water Resource Management Tools Definition and Purpose. 10

1.4      Software Utilized. 11

2.0       Primary Data Sources and Information. 12

2.1      Study Area Boundary. 12

2.2      Digital Elevation Model Data. 12

2.3      Land Cover Data. 12

2.4      Soils Data. 13

2.5      Geomorphology Data. 13

2.6      Well Data. 13

2.7      Lakes Data. 13

2.8      Streams Data. 13

2.9      Datum and Projection. 13

2.10    Model Resolution. 13

3.0       Water Management Tools Components. 14

3.1      Groundwater Susceptibility Analysis. 14

3.1.1        Overview of Drastic Model 14

3.1.2  Depth to Water 17

3.1.3        Recharge. 20

3.1.4        Aquifer Media. 23

3.1.5        Soil Media. 26

3.1.6        Topography (Slope) 31

3.1.7        Impact of Vadose Zone. 34

3.1.8        Conductivity (Hydraulic) 37

3.1.9        Results and Discussion. 40

3.2      Surface Water Susceptibility Analysis. 42

3.2.1        Overview of Surface Water Susceptibility Analysis. 42

3.2.2        Slope Factor Classification and Rating. 44

3.2.3        Distance to Water Factor Classification and Rating. 46

3.2.4        Land Cover Factor Classification and Rating. 48

3.2.5        Soil Factor Classification and Rating. 50

3.2.6        Results and Discussion. 52

3.3      Rational Method Runoff Calculator. 55

3.3.1        Overview of the Rational Method. 55

3.3.2        Overview AND PROCEDURES of the Basins Extension. 55

3.3.3        OVERVIEW AND Procedures for Rational Method Runoff CalculatOR EXTENSION.. 56

3.4      Overview of the Aquifer Probability Coverage. 58

3.4.1        Procedures For Producing Aquifer Probability Coverage. 58

3.4.2        Results and Discussion. 58

3.5      Groundwater Flow Model 59

3.5.1        Procedures for Building the Groundwater Flow Model 60

3.5.2        Results and Discussion. 62

3.6      Water Budget Analysis. 63

4.0       Conclusion. 64

5.0       Maintenance. 65

6.0       Recommendations. 66

7.0       References. 67

8.0       Appendix A – Water budget analysis. 69

Hydrologic Budget for Tenmile and Birch lakes. 70

9.0       Appendix B – BASIN1 EXTENSION DOCUMENTATION.. 80

 

List of Tables

Table 1, Primary Data Sources

Table 2, Weights assigned to DRASTIC parameters         

Table 3, DRASTIC Ranges and Ratings for Depth to Ground Water

Table 4, Ten-Mile Study Area Depth to Water Parameter

Table 5, DRASTIC Ranges and Ratings for Recharge (Net)

Table 6, Ten-Mile Study Area Recharge Parameter

Table 7, DRASTIC Ranges and Ratings for Aquifer Media

Table 8, Ten-Mile Study Area Aquifer Media Parameter

Table 9, DRASTIC Ranges and Ratings for Soil Media

Table 10, Ten-Mile Study Area Soil Media Parameter

Table 11, DRASTIC Ranges and Ratings for Topography

Table 12, Ten-Mile Study Area Topography Parameter

Table13, DRASTIC Ranges and Ratings for Impact of Vadose Zone Media

Table 14, Ten-Mile Study Area Impact of Vadose Zone Parameter

Table 15, DRASTIC Ranges and Ratings for Conductivity

Table 16, Ten-Mile Study Area Conductivity Parameter

Table 17, Results of Groundwater Sensitivity Assessment

Table18, Classification of Groundwater Sensitivity

Table 19, Slope Factor Classification and Rating                                                               

Table 20, Distance to Water Factor Classification and Rating                                 

Table 21, Land Cover Factor Classification and Rating                                         

Table 22, Hydrologic Soil Group Characteristics                                                  

Table 23, Soil Factor Classification and Rating                                                      

 

List of Figures

Figure 1, General Project Location Map                                                                          

Figure 2, Water Resource Management Tools Schematic   

Figure 3, Groundwater Susceptibility Analysis Flowchart

Figure 4, Depth to Water Map

Figure 5, Recharge Map

Figure 6, Aquifer Media Map

Figure 7, Soil Media Map

Figure 8, Topography (Slope) Map

Figure 9, Impact of Vadose Zone Map

Figure 10, Conductivity (Hydraulic) Map

Figure 11, Groundwater Sensitivity Map

Figure 12, Surface Water Susceptibility Analysis Flowchart 

Figure 13, Slope Map                                                                

Figure 14, Distance to Surface Water Map                                                           

Figure 15, Land Cover Map                                                                   

Figure 16, Soil Map                                                                               

Figure 17, Surface Water Susceptibility Map         

In support of the contract between the University of Minnesota, Duluth (UMD) Geological Sciences Department and Cass County, UMD is pleased to submit the Ten Mile Lake Association Water Resource Management Tools (WRMT).  This section of the report provides background information on the project location, personnel, definition and purpose of the tools and software utilized.

1.1          Project Location

The project location encompasses an area of approximately 92,466 acres and lies in portions of Cass and Hubbard counties, in north central Minnesota. Figure 1, General Project Location Map, depicts the location of the study area boundary and is presented on next page.  The study area is located in an area known as the Itasca/St. Croix moraine Interlobate area. 

 

                                              

1.2          Project Personnel

The following persons were involved in the production of this report.

·         Howard Mooers 

·         Dave Stark

·         Stacey Stark

·         Sue Hattenberger

·         Brennan T Mears

1.3          Water Resource Management Tools Definition and Purpose     

The WRMT were designed as a series of calculations, GIS analyses, and finally a groundwater flow model to assist with water resource management in Cass County and specifically to address issues in the Ten Mile Lake and Birch watersheds.  Figure 2, Water Resource Management Tools Schematic, is presented below and identifies the major components of this report. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


Each of the six components are defined below and addressed in individual sections in this report.

 

Groundwater Susceptibility Analysis (GWSA) – GWSA utilized the DRASTIC method to determine the susceptibility of specific areas to groundwater pollution potential.  DRASTIC is an acronym describing seven parameters controlling ground water pollution potential.  The seven parameters include Depth to ground water, Recharge (Net), Aquifer Media, Soil Media, Topography (slope), Impact of the Vadose Zone Media, and Conductivity (Hydraulic) of the Aquifer.  Rating factors and weights are assigned to each variable to determine the overall score.

 

Surface Water Susceptibility Analysis (SWSA) - SWSA describes the intrinsic factors of slope, distance to water, soil type and land cover and rates them in regards to their ability to produce surface water runoff that may lead to contamination in surface water bodies.   As with the GWSA, different runoff potential ratings and weighting factors are assigned to the variables to distinguish higher or lower runoff potential.

 

Rational Method Runoff Calculator (RMRC) – RMRC is series of tools used to determine peak discharge from a watershed.  With the provided input grids, watersheds can be calculated utilizing an Arcview extension.  Following this, a peak discharge is calculated for a specific rainfall event. 

 

Aquifer Probability Coverage (APC) –  APC is a map that describes the probability that any point within the study area is an aquifer. To develop an APC, stratigraphic information from 489 wells in the study area was extracted from CWI (file: Cass_well_raw_data.txt).  Elevations of all wells were determined from the 30 meter DEM.  The well data were imported to ArcView and the data table sorted by lithologic type.  All types deemed to be aquifer were coded with an indicator of 1, whereas all lithologic types deemed to be aquitards were coded with an indicator of 0.  Stratigraphic data for each 5-foot interval was extracted.  A total of 1244 stratigraphic intervals were coded.  Elevations and thickness of  each unit was converted to meters for geostatistical analysis.   Geostatistical analyses were completed using GSLib (Geostatistical Software Library and User's Guide by Clayton Deutsch and André Journel, 1992, 340 pp).  Vertical and horizontal variogram analysis was performed and the final results Kriged to a resolution of 100 x 100 meters horizontally and to variable thicknesess vertically.  The results were output to a 3D ASCII grid file for import to GIS software.

 

Groundwater Flow Model (GWFM) -  The purpose of a numerical model of groundwater flow is to provide a quantitative tool for groundwater flow in the Ten Mile Lake watershed and surrounding area.  The model is steady state, meaning it does not take into account time-dependent flow. It is regional in scale, but can be modified in the future to analyze site-specific applications.  Modeling was done using Groundwater Modeling System (GMS) developed by the Department of Defense.  GMS is a pre- and post-processor for MODFLOW, the groundwater flow model developed by the US Geological Survey. 

                                                                                                               

Water Budget Analysis (WBA) –  A hydrologic budget specifies inputs, outputs and storage changes over a reference period for a specific area.  The goal of the WBA was to examine whether there is significant exchange of water between Tenmile and Birch Lakes.  The simplest form of a hydrologic budget specifies inputs, outputs and storage changes over a reference period for a specific area.  The reference area for this investigation is the lake surface, the reference period is one year, and the study was done for the water years (Oct. 1 – Sept. 30) 2000, 2001, and 2002.  To most effectively determine the components of inflow and outflow to the lakes three separate hydrologic budgets were calculated.  First the watersheds of Tenmile and Birch Lakes were combined and an overall hydrologic budget was determined.  The relatively large size of the watersheds and the good record of stream outflow of the Boy River at Hackensack make it easier to calibrate coefficients used for determining surface runoff and evapotranspiration.  Over a one-year period the hydrologic budget should be relatively balanced with little change in storage.

 

These tools collectively present a variety of means to evaluate water resources within the study area boundary.  The GWSA and SWSA identify the intrinsic characteristics of the landscape that could lead to increased likelihood of surface or groundwater contamination.  The WBA and APC provided a framework for evaluating the groundwater flow from the Ten Mile to the Birch Lake watersheds and provided inputs for the GWFM.  The GWFM and RMRC can be utilized to predict surface and groundwater direction of flow and output.   

1.4          Software Utilized

The software utilized for this project included the following:

 

  • Arcview Version 3.2
  • Groundwater Modeling System (GMS) Version 5.0
  • MODFLOW 2000
  • Microsoft Word, Excel, PowerPoint

                       

The WMRT utilized a variety of GIS and other data sources for producing the outputs of the individual analyses.  Table 1, Primary Data Sources lists the data that was acquired or provided by Cass County, the scale of the data and a description of how it was utilized in the analysis.  Following the table the sections describe the data in greater detail.

 

 

Table 1, Primary Data Sources

 

Type of Data
Scale of Data
Data Utilization

Study Area Boundary

N.A.

All data clipped to this boundary.

Digital Elevation Model Data

1:24,000

Used for slope analysis in the GWSA and SWSA.  Utilized for inputs for deriving flow direction, flow accumulation and watersheds in RMRC.

Land Cover Data

1:100,000

Used for SWSA for coding land use with runoff potential ratings and in RMRC for coding runoff coefficients.

Soils Data

1:24,000

Used for inputs to both the GWSA and SWSA.

Geomorphology Data

1:100,000

Used for multiple inputs for the GWSA and the GWFM.

Well Data

Site specific

Used to develop the depth to water map for GWSA and for heads for the GWFM.

Lakes Data

1:24,000

Used for distance analysis in the SWSA and for general mapping purposes.

Streams Data

1:24,000

Used for distance analysis in the SWSA and for general mapping purposes.

2.1          Study Area Boundary

The study area boundaries for the WRMT were based on a shapefile produced by UMD. The name of this shapefile is Study_area.shp.  This shapefile incorporates all of the contiguous land areas within the Ten Mile and Birch watersheds.  In addition, the additional distance outside of the formal area of the Ten Mile Lakes area was needed for better definition of the regional groundwater regime being modeled in the GWFM.

2.2         Digital Elevation Model Data       

A 30-meter digital elevation model (DEM) was acquired from the National Elevation Data Set, which was accessed at the following web page http://gisdata.usgs.gov/NED/default.asp.   This data set includes seamless elevation data for the United States and was acquired at a resolution of 30 meter grid spacing.   This data set was clipped to the study area boundary defined by the file entitled Study_area.shp.  The name of the resulting DEM is Dem_nad83. 

2.3         Land Cover Data     

Land cover data was provided by DNR from their Data Deli (http://deli.dnr.state.mn.us, LandSat-Based Land Use-Land Cover (Vector)) and was also clipped to the study area boundary and was used for coding runoff potential ratings in the SWSA and for coding runoff coefficients for the RMRC. The name of the resulting shapefile is landcov.shp.                       

2.4         Soils Data

Soils data was provided by Cass County Environmental Services  and was also clipped to the study area boundary and was used for coding runoff potential ratings in the SWSA.   The name of the resulting shapefile is soilstenmile.shp. 

2.5         Geomorphology Data

Geomorphology data was acquired from the Minnesota Data Deli, which is accessed at http://deli.dnr.state.mn.us/.  This data was originally compiled under the direction of Dr. Howard Mooers and was utilized for components of the GWSA, APC and GWFM.  This data was also clipped to the study area boundary and the name of the resulting shapefile is geomorph.shp.

2.6         Well Data

Well data was acquired from the County Well Index maintained by the Minnesota Department of Health.  The well data was used in combination with the elevation of lakes in the study area to produce a depth to water map.  This map was utilized in the GWSA and the GWFM. This data was also clipped to the study area boundary and the name of the resulting shapefile is Wells.shp.

2.7         Lakes Data

Lakes data was acquired from the Minnesota Data Deli, which is accessed at http://deli.dnr.state.mn.us/.  The data was utilized in the distance analysis in the SWSA and for general mapping purposes.  This data was also clipped to the study area boundary and the name of the resulting shapefile is lakes.shp.

2.8         Streams Data  

Streams data was acquired from the Minnesota Data Deli, which is accessed at http://deli.dnr.state.mn.us/.  The data was utilized in the distance analysis in the SWSA and for general mapping purposes.  This data was also clipped to the study area boundary and the name of the resulting shapefile is rivers.shp.

2.9         Datum and Projection

All data was either acquired or converted to Universal Transverse Mercator (UTM) North American Datum (NAD) 83.  The standard unit of measure for this datum is meters.  Distance analysis was performed using feet as the unit of measure.

2.10       Model Resolution

The model resolution for the individual analyses are listed below:

 

Groundwater Susceptibility Analysis (GWSA) – 30 meter

Surface Water Susceptibility Analysis (SWSA) - 30 meter

Rational Method Runoff Calculator (RMRC) - 30 meter

Aquifer Probability Coverage (APC) - 100 meter horizontal and 2-6 meters vertical

Groundwater Flow Model (GWFM) - 100 meter horizontal and 2-6 meters vertical

Water Budget Analysis (WBA) – N/A Water budget scale is described in the report in Apendix 1.

                                                         

                                                                                                                                                           

                                   

3.1          Groundwater Susceptibility Analysis

DRASTIC is a standardized methodology used to evaluate the potential for ground water pollution potential in hydrogeologic settings (Aller et al. 1987).  A panel of managers, scientists, and private consultants developed the method. The panel included individuals representing federal, state, and local agencies, the Canadian government, and private industry.  Through a series of discussions, technical applications, and scientific reviews the panel developed what has become one of the most commonly used methods to evaluate ground water pollution potential in the United States (USEPA 1995).

 

The DRASTIC method was developed within the framework of the existing classification system of ground water regions of the United States.  Using this classification system it is possible to subdivide each ground water region into hydrogeologic settings  d on locally specific ground water characteristics.  A hydrogeologic setting is defined as a composite description of the major geologic and hydrologic factors, which affect and control ground water movement into, through, and out of an area (Aller et al. 1987).  The DRASTIC method is  d on the concept of hydrogeologic settings and is the acronym describing seven parameters controlling the ground water pollution potential of a specific hydrogeologic setting.  The seven parameters include:

 

Depth to ground water,

Recharge (Net),

Aquifer Media,

Soil Media,

Topography (slope),

 Impact of the Vadose Zone Media, and

Conductivity (Hydraulic) of the Aquifer. 

 

While these parameters do not include the infinite number of variables that can be used to describe the physical characteristics of a hydrogeologic setting they are considered the most important parameters for which data are available, and for assessing the ground water pollution potential of an area.

 

DRASTIC uses a numerical ranking system to assign a relative index of aquifer sensitivity (IAS) based on the following equation (Aller et al. 1987):

 

IAS = Dw*Dr + Rw*Rr + Aw*Ar + Sw*Sr + Tw*Tr + Iw*Ir + Cw*Cr                  

 

where w and r are weights and ratings assigned to each parameter. 

 

The weights assigned to each parameter are constant, ranging from 1 to 5, and based on the relative importance in evaluating ground water pollution potential as determined by the panel through a consensus approach.  Table 2, Weights Assigned to DRASTIC Parameters is located below.  In essence, the more important a variable is considered to be in evaluating ground water pollution potential the higher its weight will be.  

 

 

 

 

Table 2, Weights Assigned to DRASTIC Parameters

DRASTIC Parameter

Weight (relative importance)

Depth to ground water

5

Recharge (net)

4

Aquifer Media

3

Soil Media

2

Topography

1

Impact of vadose zone media

5

Conductivity

3

 

Numerical rating values for each of the parameters vary from 1 to 10, and are assigned using a range of values obtained by defining the physical characteristics of each parameter within the hydrogeologic setting.  The range of values represents data derived through either consulting existing sources of hydrogeologic information, or through conducting field-sampling programs.  Rating values for D, R, S, T, and C are assigned one value per range, whereas rating values for A and I are assigned a typical rating selected from a set of variable ratings.  However, the ratings for each parameter can be adjusted  based on specific knowledge of the hydrogeologic setting in question tempered by sound professional judgment.

 

 

Methods used to derive each factor are described in the following section of the report.   Figure 3, Groundwater Susceptibility Flowchart graphically displays the inputs, data assigned and analysis steps and is presented on the next page.


 

Figure 3, Groundwater Susceptibility Analysis Flowchart

 

Assign Values

 

 

 

 
 


Rasterize Files

 

 

 

Input Themes

 

 

 

Results

 

 

 
           

 

 

 

 

 

 

 

 

 

 

 


Soil texture field of geomorphology shapefile used to create coverage.

 

 

Aquifer_

media

Values

(3-8)

 

 

Aquifer_media.shp

 

Aquifer Media

Geomorph.shp

 
                               


An aquifer is a geologic unit that can store and transmit water at rates fast enough to supply reasonable amounts to wells (Fetter 1994).  In simpler terms, an aquifer represents a geologic unit in which all the pore spaces are completed filled (saturated) with water.  Ground water within an aquifer occurs in confined, unconfined, or semi-confined conditions.  Therefore, one must take care when selecting a value for depth to water based on the characteristics of the aquifer. 

 

In a confined aquifer ground water is generally under pressure; therefore, the elevation of ground water observed in a well can be higher than the elevation of the water table beneath the confining layer.  In this case, depth to water should be measured at the top of the aquifer, which also corresponds to the base of the confining layer.  Depth to water in a confined aquifer can be obtained by consulting geologic reports containing maps, cross sections, or well logs.

 

In an unconfined aquifer the water table represents the expression of the surface below ground level where the pores spaces are completed saturated.  In this case, the water table is able to rise and fall under atmospheric pressure.  An unconfined aquifer can be present in any type of geologic media and may be seasonal or permanent in nature.  However, for the purposes of DRASTIC an unconfined aquifer is chosen as the depth to water table in a geologic unit that yields significant enough quantities of water to be considered an aquifer.

 

A semi-confined aquifer refers to aquifers that are overlain by a less permeable unit that restricts or retards the flow into or out of the aquifer.  Semi-confined aquifers exhibit characteristics ranging from confined to unconfined; therefore, the choice of depth to water is determined by evaluating which characteristic of the aquifer is most dominant and then follow the procedures outlined above. 

 

DRASTIC was designed for the evaluation of unconfined aquifers.  The ranges and ratings for depth to water are based on what are considered to be depths where the potential for ground water contamination significantly changes.  Table 3, DRASTIC Ranges and Ratings for Depth to Groundwater is presented below.  In cases where the depth to ground water is shallow the travel time for a contaminant released at the surface is shorter than ground water occurring at deeper levels.  Moreover, the potential for attenuation of a contaminant increases as depth to water increases.  These criteria are reflected in the assignment of ratings for the depth to water parameter.

 

 

Table 3, DRASTIC Ranges and Ratings for Depth to Groundwater

Range – Depth to Water (Feet)

Rating

0-5

10

6-15

9

16-30

7

31-50

5

51-75

3

76-100

2

100+

1

 

 

Depth to water in the study area ranges from zero feet to approximately 100 feet.  The calculated values for depth to water using DRASTIC range from a low of 5 to a high of 50.   These values are reflective of the variability of the hydrogeologic setting and overall characteristics of ground water flow in the study area.   Table 4, Ten Mile Study Area Depth to Water Parameter is presented below.

 

 

 

Table 4, Ten Mile Study Area Depth to Water Parameter

Range (FT)

Weight

Rating

Calculated DRASTIC Value

0-5

5

10

50

6-15

5

9

45

16-30

5

7

35

31-50

5

5

25

51-75

5

3

15

76-100

5

2

10

100+

5

1

5

 

A depth-to-water coverage of the study area was developed by using the elevations of surface water bodies, contouring those values and then subtracting the contoured water table elevations from the land surface elevation.  The depth-to-water grid was produced at the same resolution as the DEM, 30 meters. The resulting grid was used for assigning DRASTIC weight and rating factors and calculation of the depth to water parameter.    Figure 4, Depth to Water Map and the associated DRASTIC values are presented on the next page.


 

 

The primary source of ground water recharge is precipitation that infiltrates through the land surface and percolates into the aquifer.   The amount of water that recharges an unconfined aquifer is dependent upon three major factors: 1) the amount of precipitation not lost to evapotranspiration, 2) the vertical hydraulic conductivity of surficial deposits and stratigraphy of the unsaturated zone, and 3) the transmissivity of the aquifer and potentiometric gradient of ground water flow (Fetter 1994:512).  In a confined aquifer recharge occurs in areas where the confining layer is absent or a leaky confining layer is present.  Recharge may occur through down-flow from a higher aquifer, or through up-flow from a lower aquifer.    

 

In the DRASTIC model, net recharge is defined as the average annual amount of water that penetrates the ground surface and infiltrates to reach the aquifer.  However, it is a difficult parameter to measure and any quantification of aquifer recharge must be considered an estimate and not an exact measured value (Korkmaz 1990).  As such, the ranges and ratings used in DRASTIC provide some leeway for choosing values that are representative of the recharge for a given study area.   The amount of recharge for a given area determines the amount of water available to transport a contaminant introduced at the surface vertically to the water table and horizontally within the aquifer.  Moreover, the dispersion and dilution of a contaminant in the unsaturated zone is largely controlled by this parameter.  Table 5, DRASTIC Ranges and Ratings for Recharge (Net) are listed below.

 

 

Table 5, DRASTIC Ranges and Ratings for Recharge (Net)

Range – Net Recharge (Inches/Yr)

Rating

0-2

1

3-4

3

4-7

6

8-10

8

10+

9

 

The best estimates of recharge in the study area were calculate by St. George (1994) and range up to 12 in/yr. The calculated values for recharge using DRASTIC range from a low of 4 to a high of 24.  These values are reflective of the overall variability of the geomorphology of the region and characteristics of the vadose zone in the study area.  Table 6, Ten-Mile Study Area Recharge Parameter is listed below.

 

Table 6, Ten-Mile Study Area Recharge Parameter

Range (inches/yr)

Weight

Rating

Calculated DRASTIC Value

0-2

4

1

4

2-4

4

3

12

4-7

4

6

24

7-10

4

8

32

10+

4

9

36

 

A recharge coverage of the study area was developed using a landform- based approach to estimation of ground water recharge (St. George 1994).  As discussed earlier it is a difficult parameter to measure for a number of reasons and any quantification of recharge must be considered an estimate.  Regardless, recharge values for the study area are based on regional geomorphologic characteristics of central Minnesota.  A digital polygon geomorphology coverage of Cass and Hubbard counties compiled at a scale of 1:100,000 by UMD was used to estimate recharge.  Landforms present within the geomorphology coverage were compared to those mapped by St. George (1994), and assigned recharge values using the same procedures.  The polygon coverage was then converted to grid coverage containing estimated recharge values for Cass and Hubard Counties at a spacing of 30 meters.  The recharge grid was then clipped to the boundary of the study area, and cell values reclassified as integers.  The resulting grid was used for assigning DRASTIC weight and rating factors and calculation of the recharge parameter.  Figure 5, Recharge Map is presented on next page.


Aquifer media refers to the consolidated or unconsolidated geologic material that yields sufficient quantities of water for use.  Water is contained in aquifers within the pore spaces of clastic sediment and rock and in fractures or solution cavities within non-clastic rocks.  Aquifers that yield water from pores spaces have primary porosity, whereas aquifers that yield water from fractures or solution cavities have secondary porosity.

 

The characteristics of ground water flow in an aquifer are controlled to a great degree by the porosity of the aquifer media.  Porosity is defined as the ratio of the volume of void spaces in a geologic unit to the total volume of the geologic unit.  Clastic sedimentary geologic units generally have primary porosity that is influenced by grain size, shape, and sorting all of the clastic materials and this contributes to the arrangement or packing of grains within the unit.  Packing is important because it largely determines the amount of void spaces available for water storage.  In general, sedimentary units that are poorly sorted typically contain a wide range of grain sizes and have lower porosities compared to sedimentary units that are well sorted and contain a small range of grain sizes.  Non-clastic rocks generally have secondary porosity and water is stored in and transmitted through fractures and solution cavities within the aquifer.

 

In DRASTIC the ranges of aquifer media types are given as descriptive names with rating values listed in order of increasing pollution potential.  Table 7, DRASTIC Ranges and Ratings for Aquifer Media is presented below.  The relative pollution potential of each media type is based on information obtained from observations made from studies conducted in various hydrogeologic settings.  The method allows for flexibility in selected a rating value based on professional expertise or specialized knowledge of the aquifer media present within a given study area.   

 

Table 7, DRASTIC Ranges and Ratings for Aquifer Media

Range – Aquifer Media

Rating

Typical Rating

Massive Shale

1-3

2

Crystalline Rock

2-5

3

Weathered Crystalline Rock

3-5

4

Glacial Till

4-6

5

Bedded Sedimentary Rock Sequences

5-9

6

Massive Sandstone

4-9

6

Massive Limestone

4-9

6

Sand and Gravel

4-9

8

Basalt

2-10

9

Karst Limestone

9-10

9

 

As a whole the aquifer media is dominated by the mixed sediments of the Itasca and St. Croix moraines.  The calculated values of the aquifer media parameter using DRASTIC range from a low of 3 to a high of 8.  Table 8, Ten Mile Study Area Aquifer Media Parameter is listed below.  These values reflect the complexity of the subsurface geology of the study area. 

 

Table 8, Ten Mile Study Area Aquifer Media Parameter

Aquifer Media Range

Weight

Rating

Calculated DRASTIC Value

Superglacial

3

4-6

5

Outwash*

3

5-7

6

Ice Contact

3

7-9

8

Till

3

2-4

3

*Note: a few small wetlands were included within outwash polygons and were coded as outwash.

 

An aquifer media coverage of the study area was developed using data derived from the geomorphology coverage of Minnesota.  The aquifer media types of the study area was compiled into a point coverage with a spaceing of 30 meters.  The point coverage was then converted to a grid of aquifer media that was clipped to the boundary of the study area.  The coverage was reclassified  based on lithology.  The resulting grid was used for assigning DRASTIC weight and rating factors and calculation of the aquifer media parameter.  Figure 6, Aquifer Media Map is presented on the next page.


 

Soil media refers to the uppermost weathered zone of the earth, which typically extends from the land surface to an average depth of 60 inches.  Soil formation is a complex process where the interaction and influence of climate, organisms, and topographic factors acting on the soil parent materials over time result in the development of a soil profile.  The soil profile contains a number of diagnostic surface and subsurface horizons that are classified on the basis of quantifiable physical and chemical criteria.  The genetic horizons potentially developed within a soil profile are typically arranged in the following sequence the O, A, E, B, C and R horizons (Buol et al. 1997).  There are a number of other potential arrangements and combinations of soil horizons; however, for the purposes of this project only the aforementioned horizons will be discussed.    

 

The O horizon is a generally associated with organic soils and is characterized as a soil layer dominated by organic materials formed or deposited on either an organic or mineral surface.  The A, E, B, C, and R horizons are associated with mineral soils. 

 

The surface A horizon is a soil layer formed at the surface or below an O horizon.  It is characterized by the accumulation of organic matter derived from the decay of plant and animal tissue, and various humic compounds.  Surface A horizons vary in thickness depending on the factors involved is soil genesis, but are generally thicker where grasses dominate.

 

An E (elluvial) horizon is a subsurface soil layer formed below the A horizon that is characterized by the elluviation or loss of clay, iron, aluminum and other compounds resulting in a concentration of quartz or other weathering resistant minerals in silt or sand size particles. 

 

The B (illuvial) horizon is a subsurface layer formed below the A and E horizons in which the dominant features are characterized by one or more of the following: 1) illuvial concentration of silicate clay, iron, aluminum, and other compounds alone or in combination, 2) evidence of removal of carbonates, 3) coatings on the faces of peds, 4) alteration of material from its original condition that obliterates the original rock structure, or 5) any combination of these. 

 

The C horizon is a subsurface layer that shows little evidence of alteration by soil forming processes and lack the properties of the O, A, E, and B horizons.  The C horizon represents the parent material for soil formation that may or may not be similar to the material in which the other horizons are formed.  The R horizon is a layer consisting of consolidated or incompletely weathered bedrock material.

 

Soils, when present, offer the first line of defense in the protection of an aquifer from contamination.  The soil has a significant impact on the timing and amount of water that infiltrates into the ground surface and is available for percolation to recharge the aquifer.  Moreover, the amount of organic matter present in the soil has a profound influence on the adsorption and complexation of contaminants released at or near the surface.  In DRASTIC the ranges of soil media are based on the soil textural classification chart and given ratings based primarily on grain size.  Table 9, DRASTIC Ranges and Ratings for Soil Media is presented below. In general, finer grained soils (e.g. clays, silts) have a low rating due to their ability to attenuate or slow the migration of contaminates as compared to coarse-grained soils (e.g., sands, gravels).

 

 

Table 9, DRASTIC Ranges and Ratings for Soil Media

Range – Soil Media

Rating

Thin or Absent

10

Gravel

10

Sand

9

Peat

8

Shrinking/aggregated Clay

7

Sandy loam

6

Loam

5

Silt Loam

4

Clay loam

3

Muck

2

Non-shrinking/aggregated Clay

1

 

 

 

The soils present across the study area are dominated by  sandy loam, loamy sand and muck distributed throughout the area.  The calculated values for the soil media parameter using DRASTIC range from a low of 4 to a high of 20.  Table 10, Ten Mile Study Area Soil Media Parameter is listed below.  The lower values correspond to fine textured soils that have low infiltration rates, whereas the high values correspond to coarse textured soils with high infiltration rates.    Overall the soils within the bounds of the study area are representative of the region as a whole.

 

 

 

 

 

 

Table 10, Ten-Mile Study Area Soil Media Parameter

 

Soil Name
HYDGRP
Texture
Weight
Rating
DRASTIC

Akeley-Debs

A

Loamy sand

2

0

16

Alstad

C

Fine sandy loam

2

6

12

Aqualfs

C

Clay loam

2

2

4

Arenic Eutroboralfs

B

Silty clay loam

2

2

4

Baudette

B

Silt loam

2

0

12

Bergkeller

B

Sandy loam

2

6

12

Bootlake-Graycalm

A

Sandy loam

2

0

12

Bowstring-Seelyeville

A

Muck

2

2

4

Cathro

A

Muck

2

2

4

Cathro-Seelyeville

A

Muck

2

2

4

Cromwell

A

Sandy loam

2

6

12

Cushing

B

Loam

2

5

10

Cutaway

B

Sand

2

10

20

Debs-Akeley

B

Silt loam

2

0

12

Demontreville

B

Loamy sand

2

8

16

Demontreville-Mahtomedi-Cushing

B

Loamy sand

2

8

16

Egglake

B

Loam

2

0

10

Fluvaquents

D

Sandy loam

2

6

12

Friendship

A

Sand

2

10

20

Glossaqualfs

B

Sandy loam

2

6

12

Graycalm

A

Sand

2

10

20

Graycalm-Bootlake

A

Loamy sand

2

0

16

Graycalm-Mengha

A

Loamy sand

2

0

16

Graycalm-Sanburn

A

Loamy sand

2

0

16

Greenwood

A

Peat

2

2

4

Haslie-Nidaros

D

Muck

2

0

4

Haslie-Seelyeville-Cathro

D

Muck

2

0

4

Histosols

A

Muck

2

2

4

Mahtomedi

A

Loamy sand

2

8

16

Markey

A

Muck

2

2

4

Meehan

B

Sand

2

10

20

Menahga

A

Sand

2

10

20

Menahga-Cutaway-Glossic Eutroboralfs

A

Coarse sand

2

10

20

Mooselake-Lupton

A

Mucky peat

2

0

4

Nidaros

A

Muck

2

0

4

Pits, Gravel-Udipsamments

A

Sand

2

0

20

Potatolake

B

Very fine sandy loam

2

0

14

Rifle

A

Mucky peat

2

2

4

Roscommon

A

Sand

2

10

20

Sanburn

B

Loamy sand

2

0

16

Sanburn-Graycalm

B

Loamy sand

2

0

16

Sandwick

B

Loamy sand

2

8

16

Seelyeville

A

Muck

2

2

4

Spooner

C

Silt loam

2

0

12

Steamboat-Two Inlets

B

Sandy loam

2

0

12

Steamboat-Two Inlets-Selleyeville

B

Sandy loam

2

0

12

Stuntz

C

Very fine sandy loam

2

9

18

Sugarbush-Two Inlets

B

Sandy loam

2

0

12

Two Inlets-Eagleview-Steamboat

A

Loamy sand

2

0

16

Typic Borohemists

A

Mucky peat

2

2

4

Typic Borohemists-nonacid-Typic Borosaprists

A

Mucky peat

2

2

4

Typic Borosaprists-Bowstring

A

Muck

2

2

4

Typic Udipsamments

A

Gravely coarse san

2

10

20

Typic Udipsamments-Arenic Eutroboralfs-Alfic Udipsam

B

Sand

2

10

20

Warba

B

Very fine sandy loam

2

9

18

Warba-Auic Eutroboralfs

C

Loam

2

5

10

Warba-Cromwell

B

Sandy loam

2

6

12

Watab

C

Loamy sand

2

8

16

Water

A

Water

2

2

4

Wurtsmith

A

Loamy sand

2

0

16

 

The soils coverage used in the analysis was derived from the digital soils polygon coverage provided to UMD by Cass County GIS staff.  The digital soils coverage contains polygons representing the soil map units delineated by scientists at the Natural Resources Conservation Service and the US Forest Service.  A soil map unit represents an area dominated by one or more types of soil that is identified and named according to the taxonomic classification of the dominant soil.  The polygon coverage was converted to a grid with 30 m cell size resolution using the soil map unit as the output value.  The soil grid coverage of distinct soil map units was used to assign a DRASTIC weight and rating in order to calculate the soil media parameter.  Figure 7, Soil Media Map is presented on next page.


 

Topography refers to general configuration of the land surface including its relief and the position of naturally occurring and cultural features.  In the DRASTIC model, topography refers to the percent slope of the land surface and its variability throughout a hydrogeologic setting.  Topography largely controls the potential for a contaminant to runoff or remain on the land surface long enough to allow infiltration into the subsurface to occur. 

 

Methods used to calculate percent slopes vary depending on the source consulted for making a determination.  Slope can be calculated from topographic maps by measuring the change in elevation over a distance and converting to a percent, consulting published detailed soil survey maps, or using a combination of these two methods.  However, in most cases digital elevation models (DEM) of much of the United State are available allowing the user to calculate percent and assign a range of values using GIS technology.

 

The ranges for topography in the DRASTIC model correspond to the typical ranges identified by the Soil Conservation for percent slope.  Flat lying or gently sloping surfaces have low runoff capacity and are typically associated with higher pollution potential.  In contrast, steeply sloping surfaces have high runoff capacity and are associated with low pollution potential.  This is reflected in the ratings assign for use in each slope range category.  Table 11, DRASTIC Ranges and Ratings for Topography is presented below.

 

 

Table 11, DRASTIC Ranges and Ratings for Topography

Range – Percent Slope

Rating

0-2

10

3-6

9

7-12

5

13-18

3

18+

1

 

Topography characteristics of the area vary from relatively flat glacial outwash plains to the hummocky topography of the Itasca and St. Croix moraines.  The calculated values for the topography parameter range from a low of 1 to a high of 10.   The topographic variations observed across the study area are primarily a function of the complex geomorphology and geologic history of the area.  Table 12, Ten Mile Study Area Topography Parameter is listed below.

 

Table 12, Ten Mile Study Area Topography Parameter

Slope Range (%)

Weight

Rating

Calculated DRASTIC Value

0-2

1

10

10

3-6

1

9

9

7-12

1

5

5

13-18

1

3

3

18+

1

1

1

 

The topography coverage of the study area was derived from the existing 30m USGS digital elevation model (DEM) of Cass and Hubbard County.  The DEM data consist of a regular array of elevations arranged horizontally as profiles with 30 meter spacing along and between each profile.  A percent slope grid for the county was derived directly from the DEM at a 30m cell size resolution, clipped to the boundary of the study area, and reclassified as an integer grid for use in the analysis.  Ranges of percent slope were then used to assign a DRASTIC weight and rating in order to calculate the topography parameter.  Figure 8, Topography (Slope) Map is presented below.


The vadose (unsaturated) zone is defined as the zone between the land surface and the water table where the pore spaces are partially or discontinuously saturated with water.  The pore spaces within the vadose contain water at less than atmospheric pressure as well as air and other gases.  The geologic media that constitute the vadose zone determine the potential for contamination attenuation between the  base of the soil media and top of the aquifer.  Complex physiochemical processes including biodegradation, neutralization, filtration, volatilization, and dispersion of infiltrated fluids occurs within the vadose zone.  Furthermore, the geologic media constrains the migration of fluids through the vadose zone thereby controlling the amount of surface area the fluid is in contact with and the amount of time for available for attenuation.

 

The selection of vadose zone media is dependent upon whether the aquifer is confined or unconfined.  In the case of a confined aquifer the impact of the vadose zone is characterized as having a confining layer regardless of what other geologic media types are present between the soil and top of the aquifer.  In the case of an unconfined aquifer the most significant geologic media that influences pollution potential is selected.

 

In DRASTIC the ranges of impact of vadose zone media are given as descriptive names with rating values listed in order of increasing pollution potential.  Table 13, DRASTIC Ranges and Ratings for Impact of the Vadose Zone is presented below.  The relative pollution potential of each media type is based on information obtained from observations made from studies conducted in various hydrogeologic studies.  The method allows for flexibility in selected a rating value based on professional expertise or specialized knowledge of the vadose zone media present within a given study area.

 

 

 

Table 13, DRASTIC Ranges and Ratings for Impact of the Vadose Zone

Range – Impact of Vadose Zone Media

Rating

Typical Rating

Confining Layer

1

1

Silt/Clay

2-6

3

Shale

2-5

3

Massive Limestone

2-7

6

Massive Sandstone

4-8

6

Bedded Sedimentary Rock Sequences

4-8

6

Sand and Gravel with significant

Silt and Clay

4-8

6

Crystalline Rock

2-8

4

Clean Sand and Gravel

6-9

8

Basalt

2-10

9

Karst Limestone

8-10

10

 

The vadose zone media types were compiled from the geomorphology coverage.  Geomorphic landform identification is an appropriate method of estimating vadose zone materials in glaciated landscapes. The spatial distribution of vadose media types differs slightly from that determined for the aquifer, but as a whole the vadose zone is also dominated by similar units.  The calculated values of the impact of vadose zone parameter using DRASTIC range from a low of 30 to a high of 40. As is the case with the aquifer media, the impact of vadose zone values reflects the complexity of the subsurface geology of the study area.   Table 12, Ten Mile Impact of the Vadose Zone Parameter is listed below.

 

 

 

Table 12, Ten Mile Impact of the Vadose Zone Parameter

Range – Impact of Vadose Zone Media

Weight

Rating

Calculated DRASTIC Value

Superglacial

8

7

35

Outwash*

5

8

40

Ice Contact

5

8

40

Till

5

6

30

*Note: organic soils were included within outwash polygons and were coded as outwash

 

The vadose zone media types of the study area were compiled into a point coverage initially spaced at 30 meters.  The point coverage was then converted to a grid of vadose zone media that was clipped to the boundary of the study area.  The resulting grid was used for assigning DRASTIC weight and rating factors and calculation of the impact of vadose zone parameter. Figure 9, Impact of Vadose Zone Map is presented on next page.


Hydraulic conductivity, or the coefficient of permeability, refers to the ability of an aquifer to transmit water, which in turn largely controls the rate at which ground water and any contaminant contained within the aquifer will flow under a given hydraulic gradient.  Hydraulic conductivity is dependent upon the sedimentary characteristics of the aquifer media; thereby, it is a function of the grain size, shape, sorting and packing of the aquifer materials and properties of the fluid passing through the aquifer.

 

There are a number of methods available to determine the hydraulic conductivity of an aquifer.  Values can be obtained by conducting aquifer pumping tests, consulting published hydrogeologic reports, or estimating based on the properties of the sedimentary characteristics of an aquifer (Freeze and Cherry 1979).  When possible, hydraulic conductivity values for an aquifer are obtained by conducting laboratory analysis of samples collected from drilling.  Four of the most common methods used to determine hydraulic conductivity based on sediment type include: 1)  Hazen (1893), 2) Krumbein and Monk (1942), 3) Harleman et al. (1963), and 4) Puckett et al. (1985).  Each method was designed for various applications under differing aquifer conditions; however, all are empirical methods used to estimate hydraulic conductivity based on grain size distribution of the aquifer materials.  The selection of a specific method to estimate hydraulic conductivity is dependent on the purposes of the analysis; therefore, some level of professional judgment should be exercised when assigning a value.

 

The ranges and ratings for hydraulic conductivity used in the DRASTIC model represent values where the potential for ground water contamination is considered to significantly change. However, the method allows for flexibility in selecting a range and rating value based on professional expertise or specialized knowledge of the ground water flow system.    Table 15, DRASTIC Ranges and Ratings for Conductivity is presented below.  

 

Table 15, DRASTIC Ranges and Ratings for Conductivity

Range – Conductivity (Hydraulic GPD/Ft2)

Rating

1-100

1

100-300

2

300-700

4

700-1000

6

1000-2000

8

2000+

10

 

 

The calculated values for the hydraulic conductivity parameter for use in DRASTIC range from a low of 3 to a high of 30.  Table 16, Ten Mile Study Area Conductivity Parameter is listed below.

The lower hydraulic conductivity values are associated with the lacustrine silt and glacial till whereas the higher values are associated with the lacustrine sand and outwash.  

 

 

 

 

Table 16, Ten Mile Study Area Conductivity Parameter

Range - Conductivity (GPD/FT2)

Weight

Rating

Calculated DRASTIC Value

Outwash*

3

10

30

Ice Contact

3

10

30

Superglacial

3

8

24

Till

3

1

3

*Note: organic soils were included within outwash polygons and were coded as outwash

 

 

A hydraulic conductivity coverage of the study area was derived from the geomorphology coverage based on the study of St. George (1994).   

                                                                                                               

 

Figure 10, Conductivity (Hydraulic) Map is presented on next page.


 

The results of the aquifer sensitivity assessment provide insight into the complexity of the hydrogeology and geomorphology of the study area, and are considered a reliable representation of the ground water pollution potential of the area.  Index of Aquifer Sensitivity (IAS) values calculated for the study area range from a low of 87 to a high of 206  based on 376,349 cells at 30 m resolution.  The range values were subdivided into four categories of aquifer sensitivity  based on an equal interval classification: 1) low (87 - 120), 2) moderate (120 - 160), 3) high (160 - 190), and 4) very high (190 - 206).  This classification method is commonly used as means to assign a sensitivity ranking for interpretation purposes.  The mean value of aquifer sensitivity is 164.9, which correspond to the High sensitivity classification.  The standard deviation calculated for the aquifer sensitivity assessment is 18.6.  A summary of the results derived from the aquifer sensitivity assessment of the study area based on the DRASTIC model is presented below in Table 17, Results of Groundwater Sensitivity Assessment. 

 

 

Table 17, Results of Groundwater Sensitivity Assessment

Number of Cells

Minimum

Maximum

Average

Standard Deviation

376,349

87

206

164.9

18.6

 

Table 18, Classification of Groundwater Sensitivity is presented below. 

 

Table 18, Classification of Groundwater Sensitivity

 IAS Interval

Sensitivity Classification

87 – 120

Low

120 – 160

Moderate

160 – 190

High

>190

Very High

 

A map showing the spatial distribution of aquifer sensitivity in the study area is provided below as Figure 11, Groundwater Sensitivity Map. 

 

 


 

3.2         Surface Water Susceptibility Analysis                               

Fundamentally, areas that are more prone to runoff are capable of carrying suspended sediments to water bodies.  The resultant contamination caused by these sediments is referred to as non-point source pollution. While diffuse in nature it can often greatly impair water bodies.  The surface water susceptibility analysis (SWSA) was developed to help managers identify areas that have the potential to generate higher overland flows and therefore non-point source pollution. 

 

The SWSA involved analyzing four intrinsic factors that influence surface-water runoff potential for the study area.  GIS themes were created and classified to rate four factors that affect overland flow.  The four factors are slope, distance to water (streams and lakes), land cover and soil properties.  Each factor was weighted to reflect its’ contribution to surface-water runoff and therefore a general indication of surface water contamination potential.  The factors were then combined to estimate susceptibility values for the entire study area.  These susceptibility values were classified to indicate lowest to highest potential to generate surface-water runoff. 

 

Values for the slope analysis, distance to water and soil analysis were classified and assigned a runoff-potential rating (RPR) on a scale of 1 to 10.  Values for the land cover analysis ranged from 0-9 where zeros were used to account for water or wetlands.  A rating of 1 indicates the lowest RPR for a given factor, and a 10 indicates the highest RPR for a given factor.  Each factor was then assigned a weighting score.  The RPR times the weighting factor resulted in the SWSA total score for each variable.  RPR’s and weighting scores were based on literature review and a similar susceptibility analysis completed by the USGS (USGS, 2001).  Methods used to derive each factor are described in the following section of the report.   Figure 12, Surface Water Susceptibility Flowchart graphically displays the inputs, data assigned and analysis steps and is presented on the next page.


 

Figure 12, Surface Water Susceptibility Analysis Flowchart

 

Assign Values

 

 

 

 
 


Rasterize Files

 

 

 

Input Themes

 

 

 

Results

 

 

 
           

 

Reclassify and Rename

Old value %

rpr

wf

total

<=2

1

3

3

>2-<=5

3

3

9

>5-<=10

5

3

15

>10-<=20

8

3

24

>20

10

3

30

 

 

 
 

 


Slope Analysis

 

Derive slope in % using DEMAT extension, Horn method 8 neighboring cells.

Rename.

 

 
p

 

 

 

 

 

 


                               


 

 


Slope Analysis

 

Slopes were derived utilizing the Demat.avx extension and the input theme Dem_nad83.  This extension was used to obtain a slope value in percent according to standard procedures (Burrough et al, 1998).  Slope is used as an indicator of potential runoff or infiltration of precipitation.  Susceptibility to surface-water contamination at a given point is greater when infiltration is low and runoff is high.  Flat terrain (low-percent slope) indicates areas of low runoff and high infiltration potential.  Steep terrain (high-percent slope) indicates areas of high runoff and low infiltration potential.  Contamination potential ratings were assigned to categories so that lowest slopes received the lowest rating and highest slopes received the highest rating.  Table 19, Slope Factor Classification and Rating is presented below.

 

Table 19, Slope Factor Classification and Rating

Slope Value (Percent)

Runoff Potential Rating (RPR)

Weighting Factor (WF)

SWSA Total

(RPR * WF)

<=2

1

3

3

>2-<=5

3

3

9

>5-<=10

5

3

15

>10-<=20

8

3

24

>20

10

3

30

 

The final map results are on the next page.

Figure 13, Slope Map


Distance to water was calculated by defining buffers around surface water features (streams and lakes) and ranged from 100 – 1600 feet.  These distances do not consider topographic features that may influence localized flow patterns but rather are the straight-line distance from a given location to a water body.  Distance to water was categorized into six classes.  Areas that were closest to water bodies received the highest rating and those furthest away received the lowest rating.  Table 20, Distance to Water Factor Classification and Rating is presented below.

 

Table 20, Distance to Water Factor Classification and Rating

Distance to Water (Feet)

Runoff Potential Rating (RPR)

Weighting Factor (WF)

SWSA Total

(RPR * WF)

> 1,600

1

3

3

> 800 - <= 1,600

3

3

9

> 400 - <= 800

5

3

15

> 200 - <= 400

7

3

21

>100 - <= 200

9

3

27

<= 100

10

3

30

 

 

The final map results are on the next page.

Figure 14, Distance to Surface Water Map

Land covers were assigned a runoff potential rating (RPR) score that is generally based on the rational runoff coefficient.  The rational runoff coefficient is applied in peak flow analysis and includes a runoff coefficient that ranges from 0 – 1 where a number close to 0 produces little or no runoff and a number closest to 1 produces complete runoff.  Published runoff coefficient values were applied to each separate land cover and multiplied by 10 to determine the RPR.  A weighting factor of 4 was applied to the land cover theme. Landcover descriptions of water and wetlands were assigned a C-Value of zero. Table 21, Land Cover Factor Classification and Rating is presented below.

 

Table 21, Land Cover Factor Classification and Rating

Land Cover

Rational Runoff Coefficient

(X)

Runoff Potential Rating (RPR)

Weighting Factor (WF)

Total

(RPR * WF)

1. Open Water/ Wetlands – Bogs, Marsh and Fens

0.00

10.00

 

0.00

4

0.00

2. Gravel Pits and Open Mines

0.05

10.00

0.50

4

2.00

3. Coniferous and Deciduous Forest

0.10

10.00

1.00

4

4.00

4. Mixedwood, Regeneration Young Forest

0.10

10.00

1.00

4

4.00

5. Shrubby Grassland

0.14

10.00

1.40

4

5.60

6. Grassland

0.15

10.00

1.50

4

6.00

7. Farmsteads and Rural Residences

0.17

10.00

1.70

4

6.80

8. Cultivated Land

0.20

10.00

2.00

4

8.00

9. Other Rural Developments

0.60

10.00

6.00

4

24.00

10. Urban/Industrial (cities & towns)

0.90

10.00

9.00

4

36.00

 

The final map results are on the next page.

Figure 15, Land Cover Map


 

The Natural Resource Conservation Service (NRCS) classifies soils into four Hydrologic Soil Groups.  These groups are primarily based on the soil’s runoff potential. The four Hydrologic Soils Groups are A, B, C and D, where A’s generally have the smallest runoff potential and Ds the greatest.  The four hydrologic groups with description and characteristics are listed below in Table 22, Hydrologic Soil Group Characteristics.

 

Table 22, Hydrologic Soil Group Characteristics

Hydrologic Soil Group

Description

Infiltration Capacity/Permeability

Leaching Potential

Runoff Potential

A

Deep, well-drained sands and gravels

High

High

Low

B

Moderately deep to deep, moderately drained, moderately fine to moderately coarse texture

Moderate

Moderate

Moderate

C

Impeding layer, or moderately fine to fine texture

Low

Low

High

D

Clay soils, soils with high water table, shallow soils over impervious layer

Very low

Very low

Very high

 

Soil hydrologic groups were considered because drainage characteristics of soils affect the movement of contaminants to surface waters. This variable was included as there is some spatial variability throughout the study area.  Soil Hydrologic groups, runoff potential rating and weighting factors were added to the soils coverage as listed in Table 23, Soil Factor Classification and Rating.

 

Table 23, Soil Factor Classification and Rating

Hydrologic Soil Group

Runoff Potential Rating (RPR)

Weighting Factor (WF)

SWSA Total

(RPR * WF)

A

1

4

4

B

4

4

16

C

8

4

32

D

10

4

40

 

The final map results are on the next page.

Figure 16, Soil Map


For the purposes of this analysis open water was defined to be the streams and lakes that lie within the study area border.  Each record of the theme was assigned a runoff potential rating (RPR) and weighting factor (WF) as described in the previous sections.  The RPR was multiplied by the weighting factor yielding a total-surface water runoff value for each theme.  These values were then rasterized to the same extent as the 30-meter DEM used to produce the slope map.  The values were added together using map calculator in Arcview and resulted in the surface water susceptibility map. The final results are included on the next page. Figure 17, Surface Water Susceptibility Map.


Natural resource managers can quickly access the final surface water susceptibility analysis to relate these factors when planning new or proposed facilities or for preserving areas that should be protected.   The analysis is a common sense approach to evaluating which areas of the study area may produce surface water runoff.  Management practices can be designed and implemented most effectively when they are based on an understanding of how watershed characteristics influence surface water runoff.  Identification of areas that are highly susceptible to runoff enables managers to prioritize these areas for monitoring, protection, or remediation.  The susceptibility analysis can be thought of as a snapshot in time.  As additional or better resolution data is incorporated into the GIS this information can be updated as appropriate.  In addition, the individual inputs to the surface water susceptibility analysis can be called on when needed.  For instance, if distance to water is a factor locating a facility, the distance to water map can be used individually.


 

 

3.3         Rational Method Runoff Calculator

The Rational Method Runoff Calculator (RMRC) is an extension developed by UMD using Arcview Version 3.2.  The purpose of RMRC is to calculate the peak discharge (Q) utilizing the Rational Method from a user-defined watershed.  This is done in conjunction with another extension named “Basin1” which uses digital elevation data to derive watersheds based on a user-defined point.   

                                                                                                        

Both of the extensions (.avx files) need to be loaded in the directory where all of your extensions reside (i.e. C:\ESRI\AV_GIS30\ARCVIEW\EXT32).  Copy the basin1.avx and rmrc.avx files from the following folder ..\Cass_TenMile_Project_Final\Rational_Method_Calculator\ARC_Project\Extensions and install on your computer.  No .apr file is associated with this tool.

The following information is taken from the Minnesota Department of Transportation Drainage Manual (Mn/DOT, 2000).  

 

The rational method is commonly used to calculate peak discharge from small drainage areas and is recommended for drainage areas up to 200 acres.  The rational formula estimates the peak rate of runoff at any location in a watershed as a function of the drainage area, a runoff coefficient and the mean rainfall intensity for a duration equal to the time of concentration.    The time of concentration is the time required for water to flow from the hydraulically most remote point of the drainage area to the point under investigation. 

 

The equation is:

 

Q = CIA

 

Where:            

Q = discharge (cubic feet per second)

C = runoff coefficient representing a ratio of runoff to rainfall (unitless)

I = rainfall intensity (inches/hour)

A = drainage area (acres)           

 

The runoff coefficients were estimated from Table 3.7 of the Drainage Manual (Mn/DOT, 2000).  These coefficients are applicable for storms of 5 to 10 year frequencies.   Therefore, only these event frequencies were used as option in the RMRC tool.  The rainfall intensities were assigned based on the Zone 2 Intensity-Duration-Frequency curves provided in Table 3.8 of the Drainage Manual (Mn/DOT, 2000) because the study area boundary is located within Zone 2.

 

The rational method is commonly used for peak flow analysis when siting culverts and other flow structures.  While many factors influence peak discharge, this method presented in GIS will give environmental planners a relatively straight forward approach to estimating runoff from existing land uses or doing projections based on changed land uses.  In order to utilize the tool in this manner, a pre and post development landuse shapefile would need to be created.

Use the Basin1 extension if you do not already have a drainage area defined.   The Basin1 Extension uses digital elevation data to derive watersheds based on a user-defined point. This tool also provides the means to generate a stream network, display a raindrop path traced from a point defined by a user, and display elevations extracted from a grid (cell value).

 

For more information on the drainage basin delineation and all of the tools in Basin1, please see the Basin1 documentation located in Appendix B, Basin1 Extension Documentation.

 

Procedures to Run Basin1 Extension:

 

  1. Load and activate the Basin1 and Spatial Analyst extensions.  Review the Basins documentation and familiarize yourself with the new buttons and tools added to your toolbar.
  2. Depress the “Initiate” button.  You will be prompted to select an elevation grid.  UMD has provided a 30m-elevation grid for this purpose named, dem_nad83.  Select ..\Cass_TenMile_Project_Final\Rational_Method_Calculator\ARC_Project\Grids\dem_nad83.   In the next dialog box, choose “yes” to add it to the View.
  3. Next, you will be prompted to choose the source of the Flow Direction grid.   This grid has already been computed, therefore select “add from a file” and in the next dialog box, select ..\Cass_TenMile_Project_Final\Rational_Method_Calculator\ARC_Project\Grids\FlowDir.  A flow direction grid assigns a value representing direction of flow from each cell area based on the elevation grid.
  4. Similarly, Choose “add from a file” in the next dialog box, then select ..\Cass_TenMile_Project_Final\Rational_Method_Calculator\ARC_Project\Grids\FlowAc for the flow accumulation grid.   The flow accumulation grid sums the number of cells flowing into each cell based on the flow direction grid.  
  5. To create a stream network based on the given flow accumulation grid, depress the “Riv” button to create a stream network.   You will be asked to supply a “threshold” value.   A small number will yield a very dense drainage network; a large number will yield a sparse network.   A threshold value of 500 was used as an example to create rivtest500.shp, which is included in the same directory structure under shapefiles.   Choose “yes” in the next dialog box to select the Strahler method of stream ordering.
  6. Now you will create a drainage basin for the point of your choice.  Select the “Bas” tool and place your cursor on a point and click once.   Choose “no” to analyze the entire grid.  Choose “yes” to derive basin characteristics.   A shapefile will be created that represents the new drainage basin.  A shapefile named “basintest.shp” was created using this process, which is included in the same directory structure under shapefiles.
  7. You can also experiment with the Raindrop and Elevation Point Surface tools at this point.  The Raindrop Tool is explained in more detail in Appendix B, but basically traces the path water would take based on the elevation data provided in the DEM.

 

 

This extension requires the basin you previously created in the above procedure or a shapefile containing a single polygon representing a drainage basin.

   When you have properly loaded the extension titled “Rational Method Runoff Calculator”, the button bar in the View interface will contain a “Q” button as shown on the left.  Q is a standard symbol for discharge in hydrology.

 

Several conditions must first be met before using the RMRC extension (steps 1-4 of procedures).  

 

Procedures to Run RMRC Extension:

  1. Set the map units for the View display to meters (the data provided have units of meters).  Do this by selecting View -> Properties from the menu.
  2. Make the theme representing your drainage area the active theme in the View by selecting it.
  3. Make the land cover theme visible (but not active) in the View.  Runoff coefficients for land cover must be assigned to the “Ratn_c” field in the landcover shapefile.   This shapefile is provided in ..\Cass_TenMile_Project_Final\Rational_Method_Calculator\ARC_Project\shapefiles\landcov.shp.
  4. Uncheck any grid data themes in your view so they are not visible.
  5. Select the “Q” button and if all conditions are met, select “yes”.
  6. Choose the land cover theme (landcov.shp) from the drop down list in the next dialog box.
  7. When you are prompted to select a rainfall event frequency choose 5 or 10 years from the drop down list.   
  8. Next you will be prompted to select rainfall duration in minutes.  The duration can be 5, 10, 15, 30, or 60 minutes.
  9. A discharge (Q) value will be calculated based on the input.   Record the results on a piece of paper as they are not saved after closing the dialog box.
  10. You will be prompted to save or delete the new shapefile that was created to calculate Q.   A new shapefile will be created called “theme1” in your working directory and will be deleted unless you select to save it.   If you select “yes” to save, you will be prompted for the directory in which to save it and you can rename it at this time.   The new shapefile has the extent of the drainage basin with the different landcover types.  A field “W_C” contains the weighted runoff coefficient for each landcover type.

 

For more information on the actual scripting code written see “rmt_main.ave” in..\Cass_TenMile_Project_Final\Rational_Method_Calculator\ARC_Project\scripts\rmt_main.ave


 

3.4         Overview of the Aquifer Probability Coverage

Geostatistical evaluation of water-well stratigraphy was used in the development of 3-D aquifer probability map. An aquifer probability map is conceptual model of the subsurface that is used to estimate the probability that a particular area is underlain by an aquifer.  There are many techniques that can be used to produce a probability map.  The method chosen for this investigation was to use the Minnesota County Well Index of well drillers’ logs as a source of subsurface data. 

 

To develop an aquifer probability map, stratigraphic information from 5,799 wells in Cass County and the surrounding area (those that had stratigraphy) was extracted from the Minnesota County Well Index (CWI).  Elevations of all wells were determined from the 30 meter DEM.  The well data were imported to ArcView and the data table sorted by lithologic type.  All types deemed to be aquifer were coded with an indicator of 1, whereas all lithologic types deemed to be aquitard were coded with an indicator of 0.  Stratigraphic data for each 5-foot interval was extracted using a computer code written by Randall Barnes of the U of M Twin Cities.  A total of 30,315 stratigraphic units were subdivided into 109,632 sub units(File: Cass Co_well_intervals.txt).  Elevations and thickness of  each unit was converted to meters for geostatistical analysis. 

 

The entire well file was imported to ArcView as an event theme. The wells were then clipped to the Ten Mile lake study area.  A total of 489 wells containing 1244 individual stratigraphic units were used as input to a geostatistical software package to interpolate these values to the study area grid.   Geostatistical analyses were completed using GSLib (Geostatistical Software Library and User's Guide by Clayton Deutsch and André Journel, 1992, 340 pp) that is incorporated into GMS.  Vertical and horizontal variogram analysis was performed and the final results Kriged to a resolution of 100 x 100 meters horizontally and cells varying from 2 – 6 meters vertically.  The values were then output as an ASCII grid file that can be imported to 3D analyst.  The output 3D ASCII grid can be found as (File: AquiferProbability.dat).

The resulting aquifer probability coverage is a conceptual representation of the potential that any location in the subsurface may be an aquifer.  The input data are either 1 or 0 representing aquifers and aquitards, respectively.   The resulting values there vary over a continuum from 0 to 1.  Often other properties such as hydraulic conductivity are derived from the probability values.  However, there are some limitations.  First, since wells are often concentrated in certain areas, the aquifer probability values are more accurate in those areas.  Predicted values far from wells should be considered somewhat unreliable.  Generally, aquifer probability values near 0 or near 1 indicate are considered most reliable.  Intermediate values of 0.4 to 0.6 are generally the least reliable.  When using aquifer probability models such as this a lower cutoff for aquifer of 0.6 is often used.  Values exceeding 0.6 provide a reasonable predictive value of aquifer presence.


 

3.5         Groundwater Flow Model

A numerical groundwater flow model is a computer representation of a natural landscape.  Groundwater flow can be described by mathematical approximations that can take into account a variety of hydrogeologic conditions.  Once set up and calibrated, they can be used to determine direction of groundwater flow, quantity of groundwater flow, location of recharge and discharge areas, and for water supply studies.  In addition, though particle trace analysis, they provide accurate delineation of Wellhead Capture Areas for wellhead protection studies

 

Numerical groundwater flow models generally employ one of these five numerical methods:  finite differences, finite elements, integrated finite differences, the boundary integral equation method, or analytic elements.  However, finite differences and finite elements are more commonly used than the other three (Anderson and Woessner, 1992).

 

Numerical groundwater flow models typically are run from a computer.  In their simplest form, a matrix of algebraic equations are created approximating the partial differential equations.  These algebraic equations are then numerically solved.  The numerical method chosen for the groundwater flow model will determine how the equations are estimated and how they are solved.  (Anderson and Woessner, 1992)  Hydraulic head is the unknown solved for in these equations.

 

For this study, MODFLOW was chosen as the modeling software program for several reasons.  These reasons include proven reliability, numerous pre/post processors and numerous particle-tracing options (i.e. MODPATH).  In addition, State Agency familiarization with MODFLOW is high thus availability of support resources.

 


MODFLOW is a block centered, finite difference, saturated flow model developed by the United States Geological Survey.  It is based on the following partial-differential equation describing three-dimensional movement of ground water through earth material:

where:

 

·         x, y, and z are cartesian coordinates aligned along the major axes of hydraulic conductivity Kxx, Kyy, Kzz

·         h is potentiometric head (Length)

·         W is a volumetric flux per unit volume and represents sources and/or sinks of water (1/Time)

·         Ss is the specific storage of the porous material (1/Length)

·         t is time (Time) (Harbaugh and McDonald, 1996).

 

The finite difference modeling procedure is relatively straightforward.  Quinn (1992) summarizes the finite difference model:  The modeling area is chosen and overlaid by a finite-difference grid.  Boundary conditions are assigned to the model according to available or inferred hydrologic features.  The choice of boundary conditions and their locations may necessitate a change in the configuration of the modeled area.  Flow equations are assigned to each cell in the model depending on aquifer type and stresses imposed on the system.  The equations are solved by one of several iterative methods.  The calculated values represent hydraulic head for each cell within the grid.

 

The groundwater flow models were constructed and analyzed using the Department of Defense Groundwater Modeling System (GMS) software version 5.0.  Along with its many other features, GMS serves as a MODFLOW pre and post processor.  The groundwater flow models were constructed with the following approach.  Conceptual models were established in GMS through importing base maps created with the project GIS using DEMs defining elevations.  Based on the conceptual models as defined through GMS, MODFLOW input files were created.  MODFLOW was run and the model calibrated by adjusting recharge within each conceptual model.  A particle trace was then performed to delineate the wellhead capture zone (wellhead protection area).

 

The base maps imported into GMS consisted of the geomorphology, surficial water features, and some general reference features (e.g. major roads).  Using GMS the base maps were georeferenced to their corresponding NAD 83, UTM coordinates.  DEMs were imported defining surficial elevations.  The registered base maps provided a guide for determining model boundaries and served as a reference for viewing post processed MODFLOW data.  GMS uses coverages, or layers of thematic data, in defining a conceptual model.  The process employs a "GIS approach" whereby points, lines and polygons, established in real world coordinates, are assigned attributes describing the physical features to be modeled.  GMS uses the following coverages in defining a conceptual model:  sources and sinks (such as wells, rivers, lakes and drains), recharge zones, and layer attributes (hydraulic conductivity, porosity, etc.).  Referencing the geomorphology and hydrologic features, polygons and arcs were established to delineate model boundaries, recharge areas, aquifers, and aquitards.  DEMs defined land surface elevations and were referenced when calculating buried surface elevations.  Once the conceptual model was defined in GMS, it was edited during the calibration process.  Based on conceptual model information, GMS created MODFLOW input files for processing.

 

The input files were loaded into MODFLOW, and processed.  Model results were accessed with GMS and head values were contoured for evaluation.  Model calibration was achieved when calculated potentiometric surfaces matched actual potentiometric surfaces within 3 to 4 meters.  Unless otherwise noted, recharge was adjusted to achieve calibration.  Hydraulic conductivity was considered a known parameter because it is directly derived from the LSA approach to terrain analysis.

Wellhead capture zones were delineated by the particle traces, the calibrated MODFLOW output files, and additional input files created by GMS.  MODPATH output files were accessed with GMS and evaluated.  The pathlines delineating the 10-year time of travel were then exported as shape files for integration into the project GIS.

3.5.1.1          Extract Aquifer Probability Coverage

The aquifer probability coverage explained in section 3.4 above was used as input to the groundwater flow model.  Aquifer probability data must now be converted to hydraulic conductivity.  This method uses the 3D aquifer probability data to assign hydraulic conductivity.  The basic idea is that if the aquifer probability value for a specific location is close to 1.0, then there is great likelihood that it represents a good aquifer that will have a high K value.  Conversely, if the aquifer probability value for a specific location is close to 0.0, then there is great likelihood that it represents a good aquitard that will have a low K value. 

 

This procedure transforms the linear aquifer probability data distributed between 0 and 1 to hydraulic conductivity on a log scale over a range that is specified by the user.  The range of K values for an area is typically known or at least one can make a reasonable guess.  Once the range of K is specified as (i.e. Kmin and Kmax) the expression

 

 

can be used to interpolate K between the high and low values specified. 

 

The range in K can be specified by the person conducting the analysis based on their knowledge of the area or based upon the estimated landform sediment assemblage hydraulic conductivity from dnr_geomoprh.shp coverage.  In this latter case, find the geomorphic unit within the study area with the highest K value (say 0.01 m/s).  Accordingly, find the LSA with the lowest K value (possibly 0.0000001 m/s).  An aquifer probability value of 0.01 will then be associated with a K=0.01 and an aquifer probability of 0.0 will be associated with K=0.0000001.  This method linearly correlates the aquifer probability value with log K.

 

3.5.1.2          GMS and MODFLOW Modeling Steps

The set up of a groundwater flow model is conceptually straight forward, however, the underlying mathematics and solution methods are complicated.  This discussion is a general procedure only.  In brief, a grid is set up and parameters are assigned to the grid.  These parameters include elevation, hydraulic properties such as conductivity and porosity, and recharge values.  The model also needs boundary conditions such as the known elevation of streams or known values of flow.

 

The first step is to develop a conceptual model.  GIS coverages of lakes and rivers provide surface boundary conditions.  The DEM provides the geometry of the surface.  The elevation of the base of the modeling domain was specified at 370 meters elevation based on limited depth-to-bedrock data in the area.  Once the conceptual model is defined it is then transformed into a finite-difference domain.

 

Define 3D grid – 3D grid can then be defined any way desired.  For this model the Grid Frame option in the Feature Objects menu was used within the Map Module.  This option allows the used to modify the grid to fit the modeling objectives.  A cell size of 100 x 100 meters was chosen.

 

Define land surface elevation – The goal is to use the 30m DEM as the source for the surface elevation and specified head boundary conditions.  To accomplish this, the dimensions of the modeling grid are used to extract the elevation data.  From within a preliminary flow model in GMS, we created a 2-D grid from the 3-D modeling grid.  The model grid was exported to Arc/View and elevations corresponding to each 100 meter cell were extracted from the 30 meter DEM.  The resulting grid was then reopened in GMS.

 

We then used the Data Calculator in GMS to define the top and bottom elevations of the layers that were defined for the modeling area.  This model utilized 10 layers to approximate the complex glacial stratigraphy determined by the aquifer probability map. The top of layer 1 was determined by the elevation DEM and the bottom of layer 10 was specified at 370 m.  Intermediate layer elevations were calculated accordingly.

 

Specify horizontal hydraulic conductivity – Aquifer probability data was then converted to hydraulic conductivity s explained above in section 3.5.1.1. 

 

Using the GMS Map Module, the model boundary conditions were established from the river elevation and lakes were represented as general head boundaries.  Recharge rates were then specified based on the geomorphology coverage.  All landforms that are composed of sand, sand and gravel, or gravel were specified with a recharge of 10cm/yr.  All other landforms, such as till, fine-grained lake sediment, etc. were specified with a recharge of 0-2 cm/yr depending on the sediment type. 

 

With all parameters entered the model can then be run.

3.5.1.3          Calibration

Once a model is set up and run the simulation must be calibrated.  Points of known water table elevations such as wells carefully selected lakes and ponds serve as calibration points.  Model parameters are then varied and the solution recalculated.  Hydraulic conductivity and recharge are the two parameters that are varied during calibration. 

Figure 18 shows the results of the calibrated flow model.  All supporting GMS and MODFLOW files are included in the directory Groundwater Flow Model.  In general, groundwater flow is from west to east.  Ten Mile Lake receives large contributions of groundwater on the north and west sides and lesser amounts on the SW.  The model suggests that there is a large component of groundwater flow across the isthmus between Ten Mile and Birch lakes.  This groundwater flow contribution from Ten Mile to Birch lake is also supported by the hydrologic budget discussed in Appendix A.  The model has numerous applications and can be applied to a variety of problems.  However, each application may require modification of the model by trained personnel.


 

3.6         Water Budget Analysis            

The Water Budget Analysis is a separate stand-alone document and is located in Appendix A, Water Budget Analysis.


 

The Water Resource Management Tools can be used to determine different factors that may affect water quality in the Ten Mile and Birch Lake watersheds.  While different approaches were taken in each analysis, the collective information presented represents information on groundwater and surface water susceptibility to contamination, surface water discharge and groundwater flow direction and relative magnitudes. 

 

Cass County environmental staff can utilize this information to further their goals of water quality protection in these watersheds. 


 

These analyses are based on current GIS data obtained from various sources and derived by UMD.  A significant effort was made to document all input variables and steps taken in each analysis.  With this information, Cass County staff will be able to update individual themes if new or better data is obtained in the future.  In addition, if a site-specific activity or area is to be altered, these data can be input into the specific analyses to determine how these changes affect the overall score of an area.  UMD staff will be available to explain or help implement these tools.


 

It is recommended that the entire data set included on the accompanying CD be copied to its’ own directory as determined by Cass County staff.  By maintaining this structure, future updates to the data will be easily accomplished.  Arcview project files (.apr) were mapped to open in any directory structure.


 

 

Aller, L., T. Bennett, J. H. Lehr, R. J. Petty, and G. Hackett, 1987.  DRASTIC: A Standardized System for Evaluating Ground Water Pollution Potential Using Hydrogeologic Settings.  United States Department of Environmental Protection 1995, Robert S. Kerr Environmental Research Laboratory, Ada, Oklahoma.  EPA/600/2-87/035, 622 pp.

 

Anderson, M. and Woessner, W.W., 1992, Applied groundwater modeling: simulation of flow and advective transport, San Diego, Academic Press, 381 p.

 

Buol, S. W., F. D. Hole, R. J. McCracken, and R. J. Southard, 1997.  Soil Genesis and Classification, 4th Edition.  Iowa State University Press, Ames, Iowa.

 

Burrough P.A. and McDonnell R.A., 1998.  Principles of Geographical Information Systems, Oxford University Press, p.192

 

Deutsch, C.V and Journel, A.G, 1992, GSLIB: geostatistical software library and user's guide, Oxford University Press,  1992, 340 p.

 

Fetter, C. W., 1994. Applied Hydrogeology, 3rd Edition.  Macmillan College Publishing Company, New York, New York, 691 pp.

 

Freeze, R. A., and J. A. Cherry, 1979.  Groundwater.  Prentice-Hall, Englewood Cliffs, New Jersey.

 

GSW, 1991.  Criteria and Guidelines for Assessing Geologic Sensitivity of Ground Water Resources in Minnesota.  Minnesota Department of Natural Resources, Division of Waters, 122 pp.

 

Harbaugh, A.W. and McDonald, M.G.  1996, User's documentation for MODFLOW-96, an update to the U.S. Geological Survey modular finite-difference ground-water flow model: U.S. Dept. of the Interior, U.S. Geological Survey open-file report ; 96-485.

 

Korkmaz, N., 1990.  The Estimation of Groundwater Recharge from Spring Hydrographs.  Hydrological Sciences, v. 35, pp. 209-217.

 

Minnesota Department of Natural Resources (MNDNR), 1991.  A Summary Report to the Legislative Commission of Minnesota Resources, Water Resources Management, Ground Water Sensitivity.  Minnesota Department of Natural Resources, Division of Waters, 10 pp.

Minnesota Department of Transportation (Mn/DOT), 2000.  Drainage Manual. 

 

Mooers, H. D., 1988. Quaternary History and Ice Dynamics of the Late Wisconsin Rainy and Superior Lobes, Central Minnesota.  Unpublished Ph.D. dissertation, University of Minnesota, Minneapolis.

 

Quinn, J.J., 1992, Modeling of groundwater flow, leachate migration, and geostatistical aspects of a portion of the Anoka sandplain, Minnesota, M.S. Thesis, University of Minnesota, 157 p.

 

St. George, L.M., 1994.  A landform-based approach to the estimation of groundwater recharge in complex glacial topography.  M.S. Thesis, University of Minnesota, 111pp.

 

United States Department of Agriculture (USDA) 1997.  Soil Survey Cass County, Minnesota.

 

United States Geogaphical Survey (USGS), 2001. Water Resources Investigations Report 01-4143. Giorgino, M.J. and Terziotti, S.  Susceptibility Index to Surface Contamination for the Little Cross Creek Watershed, Cumberland County, North Carolina.


 

8.0   Appendix A – Water budget analysis


Hydrologic Budget for Tenmile and Birch lakes

by: Howard Mooers and Sue Hattenberger

August 21, 2003

 

INTRODUCTION

 

This report presents the results of a preliminary hydrologic budget for the Tenmile and Birch lake watersheds.  The goal of this preliminary hydrologic investigation is to examine whether there is significant exchange of water between Tenmile and Birch Lakes.  The simplest form of a hydrologic budget specifies inputs, outputs and storage changes over a reference period for a specific area.  The reference area for this investigation is the lake surface, the reference period is one year, and the study was done for the water years (Oct. 1 – Sept. 30) 2000, 2001, and 2002.  To most effectively determine the components of inflow and outflow to the lakes three separate hydrologic budgets were calculated. 

 

First the watersheds of Tenmile and Birch Lakes were combined and an overall hydrologic budget was determined.  The relatively large size of the watersheds and the good record of stream outflow of the Boy River at Hackensack make it easier to calibrate coefficients used for determining surface runoff and evapotranspiration.  Over a one-year period the hydrologic budget should be relatively balanced with little change in storage. 

 

Separate hydrologic budgets were then compiled for Birch Lake and Tenmile Lake and a modest sensitivity analysis was conducted by varying input and output parameter coefficients within accepted ranges.  This analysis of individual watersheds allows identification of components of exchange between the two basins.

 

Results suggest that there is a large component of water exchange from Tenmile Lake to Birch Lake.  The ratio of surface water to groundwater flow accounting for this exchange cannot be determined at this time.  A detailed assessment of the groundwater flow systems including a groundwater flow model and placement of a stage recorder in the stream between Tenmile and Birch Lakes will help address this question.

 

HYDROLOGIC BUDGET

General Lakes Hydrology

 

Tenmile and Birch lakes occupy the headwaters of Boy River system.  The Tenmile Lake watershed has and area of 100.7 km2 and the Birch Lake watershed has an area of 23.6 km2.  Tenmile Lake drains to Birch Lake, which in turn drains the southeast along the Boy River.  Because of the relatively low apparent flow through the river between Tenmile and Birch lakes, it is suspected that the groundwater flow component between them may be large.  The geology of the watershed is a complex sequence of glacial sediment composed of relatively impermeable till and lacustrine sediment and permeable outwash gravels. 

 

Water Balance Equations - General Introduction

 

All water balance equations are based on the premise that the difference between water inflow and water outflow over a given time period for the hydrologic system of interest (be it a watershed, lake, etc.) must equal the change in water storage in that system.  That is

 

IN - OUT = ± STORAGE

 

If inputs exceed outputs there will be an increase in storage and the right side of the equation will be positive.  If outputs exceed inputs there will be a loss from storage and the left side of the equation will be negative. 

 

The exact form of the water budget equation for any given hydrologic system varies depending upon the size of the system and the length of the time period for which the water budget is being constructed.  For example, the equation for a long reference period (on the scale years to decades) and a large watershed (perhaps a major river drainage system) would contain only a few terms:

 

P - (ET + RO) = 0

 

where      P  = precipitation, ET = evapotranspiration, RO = surface runoff

 

This equation would apply for a case in which there was no significant change in total water storage in the basin (i.e. -no significant change in groundwater, lake, or river levels) over the time period for which the balance is set up.  This assumes that there is no persistent climatic change in the basin that would cause amounts of water storage to change.  Given these conditions, water entering the basin in the form of precipitation would simply equal to the amount of water leaving the basin in the form of evaporation or runoff.

 

The water balance equation for total water storage (both on the surface and in the subsurface) for a small watershed/basin over a short reference period would contain many more terms and might be as follows:

            INFLOW         OUTFLOW      STORAGE

(P + q + g + i) - (ET + e + SRO + SSRO) = ± (SS+GS+SM+DS+I)

 

INFLOW:

P = precipitation

q = direct inflow of surface water from other basins

g = direct inflow of groundwater from adjacent basins

i = water artificially imported from other basins (pipelines, etc.)

 

OUTFLOW:

ET     = evapotranspiration

e        = water artificially exported to other basins

SRO  = surface runoff leaving the basin

SSRO= subsurface flow from basin to other basins

 

STORAGE:

SS  = change in surface storage (lakes, rivers, etc)

GS  = change in water storage in groundwater system in basin

SM  = change in amount storage of water as soil moisture

DS  = depression storage (water pooling in surface depressions after precipitation event)

I   = interception by plants and vegetative cover (water sticking to plant leaves)

 

During short reference periods many transient terms exist (such as I, DS, and SM) that become runoff or evapotranspiration over longer reference periods and therefore disappear.  An additional feature to note is that for small watersheds that are part of larger basins surface water and groundwater divides do not always correspond (therefore even if q=0 it does not necessarily mean that there is no groundwater flow into the basin from adjacent basins). 

 

The forms of the water balance equation shown above represent two extremes, the first being very general and the second being fairly detailed.  For this study we will be constructing a water budget for the Tenmile and Birch watersheds on the time scale of one year and so will be using a form of the water balance equation of intermediate complexity.

Methods

For our study all inputs and output volumes will be referenced to the lake area.  Therefore lake-level change will be our primary storage term (i.e. If the lake level does not change over the reference period, then inputs and outputs balance).  We use the following balance equation,

 

[1]        LL* AL = (P* AL +P* ALW *CLW)+ GWFC*AL -(PET* AL) + (SRO*AL),

 

where

LL:       lake level

AL:          area of the lake

P:             precipitation

ALW:       area of the lake watershed

CLW:     conversion factor for the lake watershed

PET:     potential evapotranspiration

SRO:    surface runoff.

 

Note that all measurements are referenced to lake level change. We can rearrange

 

[2]        ±GWFC*AL = ±LL* AL- (P* AL +P* ALW *CLW) + (PET* AL) + (SRO*AL)

 

and

 

[3]        ±GWFC = [±LL* AL - (P* AL +P* ALW *CLW) + (PET* AL) + (SRO*AL)]/AL

 

GWFC is the groundwater flow component.  Usually groundwater inputs and outputs are the most difficult to measure or estimate.  Therefore we assume the residual to be groundwater flow.  However, we also have no way quantify the surface water exchange between Tenmile and Birch Lakes, and this residual we call GWFC can include other components as well.

 

 

 

INPUTS

Precipitation (P) (Table 1)

The primary input to the watersheds is precipitation.  Precipitation is recorded at several locations in the vicinity including Walker, Hackensack, and Nimrod.  Data were obtained from the USGS, Minnesota Climatology Working Group, NOAA, and DNR.   Precipitation records from 2000-2003 were used (Table 1).

 

 

OUTPUTS

Evapotranspiration (ET) (Table 2)

Since the lake surface area is our frame of reference, we must be able to determine the removal of water from the surface by evaporation.  Evaporation from free water surfaces can be estimated by a number of methods.  We used the Thornthwaite method, which provides an estimate of potential evapotranspiration based on average monthly temperature and climatologic indices.  This method is approximate but provides a reasonable estimate where sophisticated measurement equipment is not available.  The equation is shown below.

 

THORNTHWAITE EQUATION FOR POTENTIAL EVAPOTRANSPIRATION

PE = Potential Evapotranspiration

PEmonth = 16(10T/I)a x CF

PEannual =  S PEmonth

 

Parameter          Definition          Units

T                      Average monthly temperature     °C

i                       i = (T/5)1.514                 unitless

                        Note:  For any given month, if T<0 then i = 0;                                                                                        that is, there is no PE for a

                               month with average temperatures below 0.

I                       Sum of twelve monthly i's           unitless

a                       a = 6.751x10-7I3 - 7.711x10-5I2 +            unitless

                                    1.7921x10-2I + 0.49239

UPE     UPE = 16(10T/I)a or unadjusted            mm

                                    potential evapotranspiration

CF                    Latitude correction factor           unitless

PE        = UPE*CF       mm

 

 

 

Data were obtained from USGS, Minnesota Climatology Working Group, NOAA, DNR and the United States Naval Observatory (Table 2).

 

 

Surface runoff leaving the basin (SRO) (Table 3)

Surface runoff is recorded at the Birch Lake Dam outlet.  Data were obtained from the DNR (Table 3).

 

 

RESULTS

The results of the hydrologic budgets are tabulated as lengths of lake level change in Table 4. 

Combined Hydrologic Budget: Birch Lake and Tenmile Lake

 

When considered the Birch and Tenmile watersheds together, the hydrologic budget for the lakes basically balances.  The GWFC ranges from –0.01 feet to 0.9feet (see Table 4).  This indicates that in order for our system to balance –0.01 feet to 0.9 feet of water must enter our system from some source.  These values are negligible for the scale and resolution of this study and suggest that the range of coefficients specifies for watershed runoff and evapotranspiration are very reasonable. 

 

Hydrologic Budget: Tenmile Lake

           

Since we do not know the magnitude of surface water outflow from Tenmile Lake through the small channel to Birch Lake, this hydrologic budget was calculated two ways.  First we include the stream outflow from Birch Lake as an output from Tenmile Lake and we then consider Tenmile to have no surface water outflow.       

 

When including SRO from Birch Lake, Tenmile Lake’s hydrologic budget indicates a nearly balanced system.  The amount of GWFC (residual) is small, ranging from –0.34 feet to 1.15 feet (see Table 4).

           

When SRO from Birch Lake is excluded in the hydrologic budget for Tenmile Lake the residual (GWFC) becomes negative.  The values are again small, ranging from –0.34 feet to –1.77 feet (see Table 4).  However, this negative budget indicates  that for the hydrologic budget to balance there must be loss of water from the basin amounting to –0.34 feet to –1.77 feet of lake level.  

 

Therefore Tenmile Lake must lose water as surface water flow or groundwater flow or some combination of both.

 

 

Hydrologic Budget: Birch Lake

 

            The results of the hydrologic budget for Birch Lake are shown in Table 4 and indicate a substantial influx of water is needed to balance the system. The values for GWFC range from 2.93 feet to 5.45 feet.  These values are significant and must be accounted for.

 

 

CONCLUSIONS

The hydrologic budget for the combined Tenmile/Birch watershed essentially balances for each of the three years of record.  The hydrologic budget for the Tenmile Lake watershed alone is balanced within the uncertainty of this method and these data and methodologies, whether SRO from the Birch Lake Dam is included or not. The outflow volume of the Boy River at Hackensack is small relative to the size of Tenmile Lake.  

 

In contrast, an influx of water must occur to balance the hydrologic budget of Birch Lake.  The amount that must be added to the lake is essentially equal to the outflow at the dam on the Boy River.  That influx must come from watersheds upstream as either surface water or groundwater flow.  Calculations of the residual (GWFC) in the Birch Lake hydrologic budget are shown in Table 5.  The fact that the GWFC in the budget is the same amount as the outflow from the dam suggests that water is simply passing through the Birch Lake watershed from Tenmile Lake upstream.   Whether the water is coming through the channel and swamps from Tenmile Lake as surface water or as groundwater is difficult to determine without further analysis and will be part of the ongoing investigation.

 


TABLE 1

Monthly and Annual Precipitation (in inches and feet)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Station

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Annual

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Walker AGC (2000)

0.11

0.85

1.98

1.50

3.12

5.10

4.19

6.12

2.88

2.35

3.86

0.70

32.76

Walker AGC (2001)

0.26

2.23

0.13

4.67

4.07

3.66

2.70

3.90

2.13

3.12

2.47

0.35

29.69

Walker AGC (2002)

0.74

0.22

0.94

1.56

2.48

6.05

3.14

5.04

 

2.35

0.53

0.76

23.81

Walker AGC (2003)

0.12

0.18

0.92

0.52

3.50

4.39

n/a

n/a

n/a

n/a

n/a

n/a

n/a

Monthly Averages:

0.31

0.87

0.99

2.06

3.29

4.80

3.34

5.02

1.67

2.61

2.29

0.60

28.75

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Hackensack (2000)

0.11

0.90

1.98

1.50

3.32

5.10

4.19

6.12

2.88

2.35

3.86

0.70

33.01

Hackensack (2001)

0.27

2.23

0.13

4.57

4.05

3.66

2.70

3.90

2.13

3.12

2.47

0.35

29.58

Hackensack (2002)

1.28

0.22

0.94

1.56

2.21

6.05

3.14

5.04

2.29

2.35

0.08

0.76

25.92

Hackensack (2003)

0.12

0.18

0.92

0.52

3.50

n/a

n/a

n/a

n/a

n/a

n/a

n/a

n/a

Monthly Averages:

0.45

0.88

0.99

2.04

3.27

4.94

3.34

5.02

2.43

2.61

2.14

0.60

29.50

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Nimrod (2000)

0.39

1.26

2.30

1.55

4.25

4.80

3.20

1.79

1.38

1.16

4.07

0.88

27.03

Nimrod (2001)

0.79

1.69

0.25

4.53

3.17

4.94

3.21

1.68

2.24

3.11

2.04

0.72

28.37

Nimrod (2002)

0.18

0.26

1.39

2.54

2.40

4.32

6.70

2.76

2.51

2.44

0.40

0.43

26.33

Nimrod (2003)

0.27

0.19

0.60

2.74

3.55

3.92

5.10

n/a

n/a

n/a

n/a

n/a

n/a

Monthly Averages:

0.41

0.85

1.14

2.84

3.34

4.50

4.55

2.08

2.04

2.24

2.17

0.68

27.24

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Total Monthly Averages:

0.39

0.87